It seems that dual graphics only works for DDR3 memory GPU's - AMD r7 240 and r7 250(ddr3).
•AMD Kaveri Dual Graphics Only Works With DDR3 Memory Based Radeon R7 GPUs
It seems that dual graphics only works for DDR3 memory GPU's - AMD r7 240 and r7 250(ddr3).
•AMD Kaveri Dual Graphics Only Works With DDR3 Memory Based Radeon R7 GPUs
yup, and I think it is almost clear that the cpu is starting to become AMD's less priority. Which makes sense actually with the introduction of Mantle which is said to totally eliminate cpu bottleneck in gaming and HSA w/c combines both cpu and gpu computing to speed up tasks.
@ltyan
there seems to be some confusion w/ regards to w/c card can be crossfired w/ the apu. The previous A10-6800k can be crossfired with the HD7750 GDDR5 fine. It doesn't make any sense if can only accommodate an R7-240. Even an R7-250 is slower than the HD7750. I'd wait for actual dual graphics review just to be sure.
Last edited by Sempron Guy; 01-15-2014 at 08:47 PM.
that's correct... since dako sila'g market share sa APU, dapat, i-improve nila before the Blue Team will eat everthing in its path. exhausted na ang IP nila in terms sa CPU performance para mag sige'g apas sa Blue Team... sakto na, ning biya sa sila (temporarily or on paper) then focus more sa APU... in reality, i'm pretty sure (my wishful thinking), the Red Team is brewing the next generation CPUs... so, Blue Team, watch out... hehehe...
again, kita'ng mga consumer ang mabulahan once naa napu'y nice na competition in these 2 CPU players...
ang mga embedded APUs (those soldered into the mobo) kay nice ra ba kaayo... hubakon ug maayo ang Atom ani ug ang Via sa embedded market but, dapat mo compete sila's price sa VIA but, their nemesis, the Green Team is also in that market segment... hehehehe...
Last edited by lloyd_joy; 01-16-2014 at 06:30 AM.
Power ang gpu sa a10 pang dota2, battlefield 4 etc, hinay ra ang a8. Kung mag apu ka then apu lang pero kung plano mag add ug laing gpu then mas nindut intel + gpu.
you only consider APU if you are up for a very decent on-board graphics... para nako, i won't compare it nalang to intel in terms of IPC kay kahibaw naman gyud ta sa resulta...
as per bro lloyd, APU is a different breed... so let's keep it that way kay ma stress ra mo mag huna2x sige mog compare og duha ka butang nga not on the same playing field...
but this APU generation it truly revolutionary... i hope 2 to 3 years from now daghan nag mo adopt sa hUMA, HSA, og Mantle API... IMO, this APU generation is kinda ahead of its time...
Intel fans care the least about their igp for one. People who are serious about performance only need look at IPC and discrete gpu scaling. And where does that lead to?
Besides, Intel doesnt care much about low-margin SKUs so does their followers.
I for one am interested in apus for the promises of assymetric crossfire as it is a viable substitute for high end mini gpus in space constraints. But by itself alone (igp) is not a rewarding alternative for Intel + discrete gpu.
Last edited by keropi; 01-16-2014 at 12:27 PM.
read this review and you'll understand why i said "ahead of its time"... maybe it was released a little bit delayed but for me that's acceptable because nobody has ever done this before on an x86 platform... just imagine the need for programmer support to fully maximize this potential technology... kaveri is not perfect, far from it, but with support, this can be a very promising approach of computing...
AnandTech Portal | AMD Kaveri Review: A8-7600 and A10-7850K Tested
In the absence of vertical integration, software optimization always trails hardware availability. If you look at 2011 as the crossover year when APUs/SoCs took over the market, it's not much of a surprise that we haven't seen aggressive moves by software developers to truly leverage GPU compute. Part of the problem has been programming model, which AMD hopes to address with Kaveri and HSA. Kaveri enables a full heterogeneous unified memory architecture (hUMA), such that the integrated graphics topology can access the full breadth of memory that the CPU can, putting a 32GB enabled compute device into the hands of developers.
One of the complexities of compute is also time: getting the CPU and GPU to communicate to each other without HSA and hUMA requires an amount of overhead that is not trivial. For compute, this comes in the form of allowing the CPU and GPU to work on the same data set at the same time, effectively opening up all the compute to the same task without asynchronous calls to memory copies and expensive memory checks for coherency.
Last edited by bim27142; 01-16-2014 at 01:24 PM.
Similar Threads |
|